Am 19.08.20 um 18:32 schrieb Alecos Papadopoulos: Hi, Here is a hack from Jack. He posted it some years ago: <hansl> function matrix my_meanc(const matrix X)     matrix B = ok(X)     return sumc(B ? X : 0) ./ sumc(B) end function matrix A = {NA, 1; 2, 2; 1, 3} mA = my_meanc(A) mA </hansl> Best, Artur

Thanks Artur. But this also highlights the disadvantages of a mailing list instead of a forum, as regards search capabilities. It appears though that we do not want to change that. We should at least contemplate the creation of a blog or something, which we could call "A Hack from Jack", and start collecting there these neat gretl hacks -which often are not hacks of course, but instances of deep familiarity with logic, programming, and gretl. I know that it is not only Jack that has contributed such, but a blog title must be an instant hit - and "A Hack from Jack" rhymes greatly and is never forgotten. "Gang of Four" would be the fair title to all the guardians of gretl, but I doubt that anybody else here believes that Andy Gill is god. Was, unfortunately. Over the years each gretl user has benefited from those hacks that answered each one's few special inquiries -but if statistics are of any worth, these "special inquiries" are not so unique. With a bit of careful tagging it could turn into a great gretl-resource. Thanks again for this Hack from Jack. Alecos Papadopoulos PhD Athens University of Economics and Business web: alecospapadopoulos.wordpress.com/ skype:alecos.papadopoulos On 19/8/2020 20:15, Artur Tarassow wrote: > Am 19.08.20 um 18:32 schrieb Alecos Papadopoulos: >> Gretl 2020d, Windows 64-bit. >> >> I give the command >> >> matrix meancolumnsB  = meanc(B) >> >> The matrix "meancolumnsB " is generated as a single-row matrix as >> expected, but with "nan" in all its cells. >> >> I can see the matrix B  ok, it is a 10000 X 9 matrix, I can >> copy-paste it, export it in spreadsheet etc. >> >> What I observe is that there are 6  rows in the "B" matrix with "nan" >> values. >> >> So it appears that the meanc() function , in contrast to the mean() >> function, cannot skip missing values? >> >> If this is the case can we do something about it? The matrix B >> collects results from MLE simulations, and there are various >> instances where we are bound to get a "nan" value (the MLE did not >> converge, or the B may collect the elements of the Hessian but the >> Hessian is not computable and the algorithm drops back to OPG, etc) >> >> Note: I also run >> >> matrix meanrowsB  = meanr(B) >> >> Here, the result came out ok, with a "nan" result for the 6 rows of >> "B" that were filled with "nan" originally. But what would happen if >> in some row I would have some numbers and some "nan"? Will the >> meanr() return then a "nan" result? >> > > Hi, > > Here is a hack from Jack. He posted it some years ago: > > <hansl> > function matrix my_meanc(const matrix X) >     matrix B = ok(X) >     return sumc(B ? X : 0) ./ sumc(B) > end function > > matrix A = {NA, 1; 2, 2; 1, 3} > mA = my_meanc(A) > mA > </hansl> > > Best, > Artur > _______________________________________________ > Gretl-users mailing list -- gretl-users(a)gretlml.univpm.it > To unsubscribe send an email to gretl-users-leave(a)gretlml.univpm.it > Website: > https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/

Am 20.08.2020 um 05:20 schrieb Alecos Papadopoulos: Hi, Let me hijack Alecos' above mentioned statement. I would like to pose a more general issue. I always felt a bit unsure about whether the mailing list is really the "best" tool for communication (questions, bug reports and discussions), _and_ for advertisement of the project outside the gretl community. Subscribing to a specific mailing list is already quite a step for some people (Oh no, another account is needed!). Given that there exists https://stackoverflow.com/ which hosts two big sites related to statistics, ML and programming, I never saw an advantage managing a mailing list on its own -- but I may be wrong: 1) "stackoverflow": https://stackoverflow.com/questions/tagged/gretl (last gretl thread from 2018) 2) CrossValidated: https://stats.stackexchange.com/search?tab=newest&q=gretl (5 threads in 2020) I've seen the relevance of stackoverflow and stats.stackexchange during the last 2 years. The first place every data scientist, analyst or programmer looks for an answer to a problem is stackoverflow. I am pretty sure this is also the case for academic researcher and students. Using stackoverflow would allow us to post issues and discuss within a big statistics/ programming community. It is _the_ common place to join people from different fields and software projects. It has all functionalities like tagging threads for improving the matching, a good search engine, the possibility to vote answers to highlight good-quality replies. One gets many different statistics (number of views and votes) to get an impression of what seems being relevant for the community. Community members can collect achievements such as "reputation" and "badges" which may motivate people to stay active. For more "features" see here: https://stats.stackexchange.com/tour Don't get me wrong, I like to be active on the gretl mailing list but I think that stackoverflow may be a chance to spread the idea(s) of the gretl project. Artur

On Mon, 7 Sep 2020, Sven Schreiber wrote: Well, isnan() does the same job for matrices. But since we've backed off on the distinction between NaN and NA, missing() for matrices might as well just redirect to isnan(). That's now done in git. Allin